shifting landscape of hemophilia therapy: implications for...
TRANSCRIPT
C R I T I C A L R E V I EW
Shifting landscape of hemophilia therapy: Implications forcurrent clinical laboratory coagulation assays
Hanny Al-Samkari1,2 | Stacy E. Croteau2,3
1Center for Hematology, Massachusetts
General Hospital Cancer Center, Boston,
Massachusetts; 2Harvard Medical School,
Boston, Massachusetts; 3Boston Children’sHospital, Boston Hemophilia Center, Boston,
Massachusetts
Correspondence
Dr. Stacy E. Croteau, 300 Longwood Ave.,
BCH 3141, Boston, MA 02115.
Email: [email protected]
Clinical coagulation assays are an integral part of diagnosing and managing patients with hemo-
philia; however, in this new era of bioengineered factor products and nonfactor therapeutics,
problems have arisen with use of traditional coagulation tests for the quantification of several of
these new products. Discussion over the use of one-stage clotting assays versus chromogenic sub-
strate assays for clinical decision making and potency labeling has been ongoing for many years.
Emerging factor concentrates have heightened concern over assay selection and availability. Emici-
zumab interferes with all aPTT based assays, rendering them unreliable and potentially falsely
reassuring to the unaware provider. This review explores considerations for coagulation assays in
the clinical setting and highlights how awareness of institutional coagulation assays and potential
limitations have never been more critical for providers caring for patients with bleeding disorders.
1 | INTRODUCTION
Clinical coagulation laboratory assays serve an important role in diagnos-
ing and supporting treatment decisions for patients with bleeding disor-
ders, yet current factor assays have gone largely unchanged for several
decades. The prothrombin time (PT) and activated partial thromboplastin
time (aPTT) serve as screening tests for an underlying factor deficiency,
while specific factor activity assays by one-stage clotting assays (OSA) or
chromogenic substrate assays (CSA) are used for diagnosis andmonitoring
factor levels following infusion of clotting factor concentrates for individu-
als with hemophilia A or B, factor VIII (FVIII) or factor IX (FIX) deficiency,
respectively. Licensure of new, bioengineered factor concentrates to pro-
vide half-life extension1–5 and an emerging group of nonfactor therapeu-
tics, such as emicizumab-kxwh (Hemlibra, Roche),6,7 have heightened
awareness of the limitations of our current clinical coagulation assays and
fueled debate over the most clinically relevant assays. With these novel
therapies, and several more products currently in various stages of clinical
trial investigation,8–10 problems have arisen for some products that can-
not be measured accurately with traditional factor assays. Bioengineered
factor products have modifications distinct from endogenous coagulation
proteins for which the existing factor assays were designed.
OSA and CSA are used routinely in the U.S. and Europe for potency
labeling of factor concentrate vials as well as clinical monitoring of patients.
While both types of assays indirectly measure FVIII activity (FVIII:C), they
utilize differentmethodologies and are thus each subject to unique sources
of variability and inaccuracy. The number of commercial reagents, and
therefore the potential for interlaboratory variability, is more pronounced
for OSA in comparison to CSA. In a 2006, UK National External Quality
Assurance Scheme (NEQAS) survey of UK clinical laboratories, 25 different
aPTT reagents, 20 different FVIII-deficient plasmas, and 13 different refer-
ence plasmas were in use for OSA, in contrast to only 5 different CSA
kits.11 Discrepancy between OSA and CSA is increasingly recognized with
newer bioengineered factor products and certain F8 pathogenic variants.12
This is not a new problem, however. Emergence of the first B domain-
deleted (BDD) recombinant FVIII (rFVIII) clotting concentrate highlighted
the clinical challenges that can arise when OSA and CSA are discordant.
Potency labeling assigned by CSA was 20% higher than when assayed
using OSA. Since OSA underestimated factor activity it was recommended
to either use CSA or OSA calibrated with the B-domain deleted product.
To overcome this challenge in theU.S.market, the amount of factor protein
in each vial was increased 20%.13,14
Global hemostasis assays, such as the thrombin generation assay
(TGA), thromboelastography (TEG) [and its modification, rotational throm-
boelastometry (ROTEM)], aPTT waveform analysis, and others are not
currently in routine clinical by hematologists but in the age of novel hemo-
static agents may offer alternative assessment of hemostasis in patients
with bleeding disorders.15 Currently, there are no CLIA-approved labora-
tory assays to quantify the hemostatic impact of emicizumab, and indeed
the presence of even small quantities of emicizumab in patient plasma
interferes with all aPTT based assays rendering them unreliable. Given the
long half-life of emicizumab, 27.868.1 days, this interference can persist
Disclosures: Al-Samkari, H: Consultancy (Agios); Croteau, SE: Consultancy
(Aptevo, Bayer, Biomarin, Bioverativ, CSL-Behring, Octapharma, and Novo
Nordisk), Research support (Genentech, Pfizer, Octapharma, Baxalta, Shire,
and Spark Therapeutics).
Am J Hematol. 2018;1–9. wileyonlinelibrary.com/journal/ajh VC 2018Wiley Periodicals, Inc. | 1
Received: 10 April 2018 | Revised: 16 May 2018 | Accepted: 20 May 2018
DOI: 10.1002/ajh.25153
AJHAJH
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 1
1082 © 2018 Wiley Periodicals, Inc. wileyonlinelibrary.com/journal/ajh Am J Hematol. 2018;93:1082–1090.
for up to 6 months after emicizumab discontinuation.16 Providers caring
for hemophilia patients in the age of bioengineered factor products and
emicizumab must have a deeper understanding of the nuances of coagula-
tion laboratory testing than ever before.
Here, we review methodology and limitations of commonly-used
laboratory clotting tests for hemophilia, and how conventional and
extended half-life factor concentrates, as well as nonfactor therapies,
impact assay interpretation with particular attention to relevance when
providing clinical care.
2 | QUANTIFYING FACTOR VII I AND IXACTIVITY
Routine diagnosis and management of patients with hemophilia depend
on factor activity assays.17,18 Additionally, these assays support accu-
rate potency labeling of clotting factor concentrates.19 Assays must be
able to accurately reflect baseline FVIII or FIX activity levels to establish
the diagnosis and severity of individuals with hemophilia. Increasingly,
postinfusion factor levels in conjunction with population pharmacoki-
netic (PK) models are being employed to generate individual PK profiles
and personalize prophylaxis regimens.20,21 Accurate results are impor-
tant for quality care. OSA is the most widely used assay used for clini-
cal laboratory monitoring, potency labeling for FIX clotting factor
concentrates, and in the United States for potency labeling of FVIII
clotting factor concentrates. While CSA is the preferred test for FVIII
potency labeling in Europe, a 2013 survey revealed that over 90% of
European coagulation laboratories utilized OSA for clinical monitoring
of FVIII replacement.22 The U.S. FDA presently endorses OSA for FVIII
replacement potency labeling to complement the predominant modal-
ity used for clinical monitoring,23 though the Medical and Scientific
Advisory Council (MASAC) of the National Hemophilia Foundation has
recommended widespread adoption of CSA in clinical coagulation labo-
ratories for those circumstances in which OSA is unreliable.24
2.1 | One-stage clotting assays
The OSA was first described in 1953 by Langdell and colleagues. It is per-
formed via mixing serial dilutions of patient plasma with equal volumes of
FVIII-deficient plasma along with an aPTT reagent (phospholipid, activa-
tor).25Once recalcified, clot formation begins and the aPTT ismeasured, Fig-
ure 1.26 Because all other coagulation factors (including VWF) are present in
excess relative to FVIII (or FIX depending on the deficient plasma used),
lengthening of the clotting time correlates with lower FVIII activity level in
the patient sample.27 The reference plasma used to prepare the standard
curve should be created from pooled donor reference plasma from 50 or
more healthy donors with equal sex representation and including all blood
groups, and it should show parallelism against the World Health Organiza-
tion standard for FVIII or FIX.27 Additionally, a modified set of dilutions of
both patient and reference plasmamay be necessary to accurately measure
very low levels of FVIII or FIX; simple extrapolation of the standard curve
rangemaybe inaccurate and should not be performed.28
2.2 | Two-stage clotting assay
Although now largely superseded by CSA, the traditional two-stage clot-
ting assay may still be found in some specialized hemostasis laboratories.
First described in 1955 by Biggs and colleagues,29 the two-stage clotting
FIGURE 1 One-stage clotting assay (OSA) for FVIII activity. Standard curve is generated by measuring the time to clot formation (aPTT,seconds) of serial dilutions of reference plasma with FVIII-deficient plasma followed by addition of aPTT reagents and calcium to initiatecoagulation. The same process is carried out with patient plasma.25 The aPTT results are plotted on logarithmic-linear scale graph paper; alllines should run in parallel (parallelism). The FVIII activity of the sample is determined by identifying the dilution of the reference plasmathat results in the same aPTT result as the sample plasma at the dilution designated 100 IU/dl for the reference plasma. In modern labora-tory practice, the OSA is frequently automated. Many different aPTT reagents (multiple types of phospholipids and activators) and variousanalyzers are in routine clinical use worldwide [Color figure can be viewed at wileyonlinelibrary.com]
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 2
2 | AJHAJH AL-SAMKARI AND CROTEAUAL-SAMKARI AND CROTEAU 1083
assay is performed by dilutions of patient plasma treated with aluminum
hydroxide to adsorb prothrombin and then incubated with an activated
serum reagent containing human serum, factor V, phospholipids, and cal-
cium. This first stage of the assay results in formation of factor Xa (FXa,
prothrombinase) in proportion to the FVIII:C in the initial patient sam-
ple.22 The second stage involves subsampling this mixture into normal
plasma containing prothrombin and fibrinogen and measurement of the
clotting time, which is correlated indirectly with the FVIII:C of the original
sample.18 From a laboratory perspective, the primary advantages of the
two-stage clotting assay in comparison with the OSA are the lack of
requirement for FVIII-deficient plasma and less interlaboratory variation.
The disadvantages include complexity, lack of automation, and lack of
commercial reagents.18
2.3 | Chromogenic substrate assays
The CSA was originally developed in 1975 by Seghatchian and col-
leagues, Figure 2.31 In the first stage, which is similar methodologically
to the two-stage clotting assay, dilutions of patient plasma are incu-
bated with FIXa, factor X, phospholipids and calcium, resulting in for-
mation of FXa proportional to the FVIII activity of the patient sample.
Small amounts of bovine thrombin are typically added in the first stage
to facilitate activation of FVIII.32 In the second stage, a FXa-selective
chromogen is added. Hydrolysis of the substrate by FXa releases a
chromophore (typically p-nitroaniline). The color intensity read by a
photometer is proportional to the amount of FXa in the sample, and
therefore also proportional to the FVIII activity.30 The advantages and
disadvantages of CSA from a laboratory perspective are like those of
the two-stage clotting assay, with the exception that several commer-
cial kits are available.
2.4 | Sources of variation/error in OSA and CSA
Each of these assays is susceptible to different sources of preanalytic and
analytic errors. Those of key interest to clinicians are summarized in Table
1. The OSA is susceptible to heparin contamination and lupus anticoagu-
lants which can alter results; the CSA is not sensitive to either of these
potential sources of error.14,33 Both assays are sensitive to the presence
of direct oral anticoagulants, which result in a falsely low measurement of
FVIII activity,34 and both will report falsely low FVIII activity if the speci-
men is clotted because FVIII is consumed during clotting. OSA, but not
CSA, is susceptible to “preactivation” of FVIII that can occur due to
thrombin activation during venipuncture because of differences in the
assay methodology. In contrast to OSA, CSA has an incubation period
which allows for activation of all FVIII present making preactivation irrele-
vant.35 Preactivation of FVIII in OSA results in an overestimation of FVIII
activity. OSA is generally less accurate in the lower ranges of FVIII activ-
ity, making it less sensitive in the initial diagnosis and differentiation of
moderate and severe hemophilia A than CSA.28
2.5 | Sources of interlaboratory variation in OSA
Several variables may impact OSA measurement and result in substantial
inaccuracy or interlaboratory variation. Numerous different commercial
FVIII-deficient plasma products are available, which may be chemically-
depleted, immunodepleted, or naturally deficient; moreover, they may
contain von Willebrand Factor (VWF).36,37 Certain commercial FVIII-
FIGURE 2 Chromogenic substrate assay (CSA). In the first stage, patient plasma is incubated with FIXa, FX in excess, phospholipid,thrombin and calcium to generate FXa. In the second stage, a FXa-selective chromogenic substrate is added. Hydrolysis of the substrate byFXa releases a chromophore; color intensity read by a photometer. The FVIII activity is proportional to FXa generated and quantified bychromophore release (absorbance measured at the specified wavelength)30 [Color figure can be viewed at wileyonlinelibrary.com]
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 3
AL-SAMKARI AND CROTEAU AJHAJH | 3
The CSA was originally developed in the 1970s with subsequent modifi-
cations, Figure 2.31 In the first stage, which is similar methodologically
1084 AL-SAMKARI AND CROTEAU
deficient plasma productsmay have small residual FVIII resulting in flatten-
ing of the standard curve at low activity ranges (0–3 IU/dL). This affects
the accuracy of the test at low FVIII concentrations.36 As with a standard
aPTT assay, results of the OSAmay be impacted by the specific aPTT rea-
gent used; dozens are commercially available. Each aPTT reagent contains
a source of phospholipid and a surface activator, such as kaolin, ellagic
acid, or silica. Specific activators may cause inaccurate measurements of
the FVIII:C or FIX:C when assaying vial potency or recovery of certain
modified FVIII or FIX proteins (see Table 2).48 The buffer used for dilution
may influence results and numerous buffers are available commercially.49
Finally, the instrument (analyzer) used to assess for clotting may be optical,
measuring turbidity as clot forms, or mechanical, measuring increased vis-
cosity as clot forms, which provides yet another potential source of inter-
laboratory variation.50 Therefore, from a laboratory perspective, OSA may
be less expensive and easier to automate than CSA, but it has numerous
possible sources of inaccuracy. If several recommended OSA technical
standardization procedures are followed, such as use of FVIII-deficient
plasma with normal VWF levels and addition of albumin to all assay buf-
fers, reduction in interlaboratory variation can be achieved.33
2.6 | Potency labeling of factor concentrate vials with
OSA vs. CSA
Debate continues over the best approach for factor concentrate vial potency
labeling. Accurate potency labeling is critical for proper dosing of patients
and valuation of factor vials. Studies comparing potencies measured by OSA
and CSA often demonstrate significant discrepancy, with some demonstrat-
ing potencies measured at up to 40% higher for the OSA51,52 and others
demonstrating potencies measured at approximately 20% higher for the
CSA.33 When utilizing the OSA for labeling, the specific aPTT reagent used
forOSA testing has been found to impact themeasured potency results.23
3 | DISCREPANT OSA AND CSA RESULTSDUE TO FVII I PATHOGENIC VARIANTS
In the assessment of FVIII:C, clinically relevant discrepancy between
assays is generally considered >20%.14 If only the OSA or CSA is
performed, accurate diagnosis of hemophilia may be compromised.
Certain F8 genotypes can result in substantial FVIII:C discrepancy
between OSA and CSA measurements. FVIII activity should be assayed
by both during the initial diagnostic evaluation because approximately
one-third of patients with mild hemophilia A secondary to single point
mutations show a significant discrepancy in FVIII activity between OSA
and CSA.53,54 This discrepancy can result in up to 10% of patients with
mild hemophilia A being misdiagnosed as normal if only one of the two
assays is performed.55–57
3.1 | Point mutations at the A1-A2-A3 interfaces
Appreciation of these pathogenic variants and their resultant impact
requires a basic understanding of FVIII structure and function. FVIII is a
heterodimer composed of a heavy chain (A1, A2, and B domains) and
light chain (A3, C1, and C2 domains).58 In the process of thrombin-
catalyzed conversion of FVIII to FVIIIa, the B domain is cleaved off and
the result is a heterotrimer comprised of an A1 subunit, A2 subunit,
and A3-C1-C2 subunit. The A2 subunit is only weakly associated with
A1 and A3-C1-C2 subunits via an ionic interaction and is therefore sus-
ceptible to dissociation from the rest of the protein; this is how physio-
logic inactivation of FVIIIa occurs.58 In patients with variants at the
interfaces between where the A2 subunit interacts with the A1 and
A3-C1-C2 subunits, such as those that disrupt potential intersubunit
hydrogen bonds, Figure 3, the measured FVIII:C by OSA is over twice
that measured by CSA. This results from the longer incubation duration
of CSA, which allows for more time for unstable FVIIIa to dissociate,
resulting in lower measured FVIII activity.12,60
3.2 | Point mutations at thrombin cleavage sites
By contrast, mutations that affect activation of FVIII by thrombin or
the binding of FVIII to FIXa or to VWF result in discrepancy such that
OSA results are lower than CSA results. Any variant affecting thrombin
activation of FVIII or its physiologic binding to FIX is unlikely to be
problematic in the CSA, where thrombin and FIX have been added in
excess and additional time for necessary reactions to occur (time not
present in the short incubation period of the OSA).57
TABLE 1 Extrinsic sources of assay interference for OSA and CSA
ConsiderationAffectsOSA
AffectsCSA Comment
Heparin contamination � * CSA contains a heparin neutralizer; if a very high amount of heparin is present, it mayovercome the neutralizer and affect the assay
Lupus anticoagulant �
Direct oral anticoagulants � � Both direct factor Xa inhibitors and direct thrombin inhibitors
Emicizumab � � Even low plasma concentrations of emicizumab normalize aPTT-based assays; CSA usingnonhuman reagents does not recognize emicizumab
Preactivation of FVIII fromvenipuncture
� Results in overestimation of FVIII activity
Range of FVIII activity � Less accurate in 1–3% range
Variant FVIII proteinstructure/stability
� � Pathogenic variant-dependent; both assays used in diagnostic evaluation
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 4
4 | AJHAJH AL-SAMKARI AND CROTEAUAL-SAMKARI AND CROTEAU 1085
TABLE2
Recom
bina
ntfactorVIII
andfactorIX
prod
ucts
availablein
theUnitedStates
andreliabilityofone
-stage
clottingassays
(OSA
)an
dch
romoge
nicsubstrateassays
(CSA
)forea
ch
Rec
ombina
ntFa
ctorProdu
ctFe
atures
OSA
/CSA
Gen
erally
Reliable
Cau
tion
warranted
Commen
ts
FVIIIConc
entrates
Adv
ate(rurioctoco
galfa)
Full-leng
thFVIII,C
HO
celllin
e�
Ady
novate
(rurioctoco
galfa
pego
l)Full-leng
thFVIII,ran
dom
pegy
lation
withbran
ched
20kD
aPEG
�
Afstyla
(lono
ctoco
galfa)
Sing
le-cha
in,B-domain
trun
catedFVIII
�Consistent
unde
restim
ationofFVIII:C
byalltyp
esofaP
TTreagen
ts,F
DA
reco
mmen
dsco
nversionfactor(m
ultiplyOSA
resultby
2)38
BAY94–9
027(dam
octacogalfa
pe-
gol)
B-domainde
letedFVIII,site-directed
pegy
lationwith60kD
aPEG
�Und
erestimationofFVIII:C
may
occur
withsilica-ba
sedaP
TTreagen
ts,
man
ufacturerreco
mmen
dsellagicaP
TTreagen
tsas
thepreferred3
2
Eloctate(rFVIII-Fc,
efmoroctoco
galfa)
B-domainde
letedFVIII,fusionwith
IgG1Fcat
carboxy-terminus
�Field
stud
y39
Koge
nate
(octoco
galfa)
Full-leng
thFVIII,B
HKcelllin
e�
Kovaltry
(octoco
galfa)
Full-leng
thFVIII,B
HKcelllin
e�
Field
stud
y40
Novo
Eight
(turoctoco
galfa)
B-domaintrun
catedFVIII
�Field
stud
y41
Novo
Eight-G
P(N
8-G
P,turoctoco
galfa
pego
l)B-domaintrun
catedFVIII,
site-directedpe
gylation
with40kD
aPEG
�Und
erestimationofFVIII:C
(>30%)may
occur
withsomeellagicacid
and
silica-ba
sedaP
TTreagen
ts;ove
restim
ationofFVIII:C
withsome
chromoge
nickits
observe
d42
Nuw
iq(sim
octoco
galfa)
B-domainde
letedFVIII,
HEK-293celllin
e�
Xyn
tha/ReF
acto
AF(m
oroctoco
galfa)
B-domainde
letedFVIII
�May
unde
restim
ateFVIII:C
unless
produ
ct-spe
cificlabo
ratory
stan
dard
used
43
FIX
Conc
entrates
Ben
eFix
(nona
cogalfa)
rFIX
�
Ixinity(treno
naco
galfa)
rFIX,T
hr-148po
lymorphism
�
Rixub
is(nona
coggamma)
rFIX
�
Alprolix
(rFIX-Fc,
eftren
ona
cogalfa)
rFIX,fusionwithIgG1Fcat
carboxy-terminus
�Und
erestimationofFIX:C
was
observe
dwithsomesilica-ba
sedreagen
tsan
dco
nsistently
forkaolin
-based
aPTTreagen
ts.OSA
(allreagen
ts)
tend
edto
ove
restim
ateFIX:C
atlow
conc
entrations
44
Idelvion(rFIX-FP,a
lbutrepe
nona
cog
alfa)
rFIX,fusionwithreco
mbina
nthu
man
albu
min
atclea
vable
linke
rpe
ptide
�Und
erestimationofFIX:C
observe
dwithkaolin
-an
dellagicacid-based
aPTTreagen
ts(byup
to50%)45
Reb
inyn
(N9-G
P,n
ona
cogbe
tape
-go
l)rFIX,site-directed
pegy
lation
withbran
ched
40kD
aPEG
�Und
erestimationofFIX:C
observe
dwithkaolin
-based
aPTTreagen
ts.
Substantialove
restim
ationofFIX:C
observe
dwithsilica-ba
sedaP
TT
reagen
ts.Most
ellagicacid-based
aPTTreagen
tsaccu
rate,bu
tsome
unde
restim
ateFIX:C
46,47
Note
notalla
PTTreagen
tsha
vebe
enstud
iedforusewithea
chfactorco
ncen
trate;
magnitude
ofdiscrepa
ncybe
twee
nOSA
resultan
dex
pected
may
vary
along
therang
eofco
ncen
trates.
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 5
AL-SAMKARI AND CROTEAU AJHAJH | 51086 AL-SAMKARI AND CROTEAU
4 | NEW CLOTTING FACTORCONCENTRATES AND CLINICALCOAGULATION ASSAYS
Generally, OSA tends to accurately assess clinical recovery of plasma
derived (pd) FVIII and full-length recombinant (r) factor concentrates
but may underestimate the true FVIII:C of B-domain deleted (BDD)
rFVIII products, with OSA measurements 20%-50% lower than CSA
measurements. In lieu of using CSA for monitoring of these agents,
potential solutions for optimizing OSA include product-specific OSA
standards (as was done for ReFacto) or a conversion factor for OSA
results (as was done for Afstyla). Afstyla, a single-chain rFVIII product,
is an example of relatively predictable discrepancy between OSA and
CSA which may be overcome by applying a conversion factor. If OSA is
used, the label guidance is to multiply the FVIII activity by two. This
approach, while feasible and relatively simple, introduces another
potential source of error in clinical measurement.61,62
Several manufacturers have engineered modifications to the FVIII
or FIX protein to slow clearance and thereby potentially reduce fre-
quency of needed infusion for prophylaxis.63 Successful prolongation
of the half-life for rFVIII has been achieved via linkage with polyethyl-
ene glycol (PEG) and the human immunoglobulin G fragment
crystallizable (Fc) region.1,2 Similar technology, glycopegylation, Fc-
fusion, and albumin fusion, has been more successful with rFIX.3–5 Use
of the OSA to assay extended half-life FVIII and FIX products has
product-specific and OSA reagent-specific challenges, Table 2.
5 | EMICIZUMAB AND COAGULATIONTESTING
Emicizumab is a recombinant, humanized bispecific monoclonal anti-
body which serves the function of activated FVIII, bringing together
FIXa and factor X (FX). Emicizumab is currently approved for bleeding
prophylaxis in hemophilia A patients with inhibitors by the FDA, EMA,
Japan and several other countries.6,7 It is administered subcutaneously
once weekly with a longer dosing interval presently under investiga-
tion. Considerations for dosing in an acute bleed setting or peri-
operatively are also being explored as is safety and efficacy of emicizu-
mab prophylaxis in hemophilia A patients without inhibitors.64
5.1 | Conversion of emicizumab plasma
concentrations to “equivalent” factor VIII activitymeasurements
While emicizumab represents a major advancement in the management
of patients with hemophilia complicated by inhibitors, coagulation labo-
ratory testing methods have not yet caught up with this novel therapy.
The ability of current clinical assays to quantify the hemostatic poten-
tial of an individual on emicizumab prophylaxis is limited. A conversion
factor of emicizumab (mg/mL) to “equivalent” FVIII:C of approximately
0.3, such that a concentration of 33 mg/mL of emicizumab would be
expected to correspond to a FVIII:C of approximately 10% based on
cynomolgus monkey studies has been suggested65 and subsequently
used in human emicizumab studies to inform initial dose selection;
however, correlative human studies are not available.66
5.2 | Limitations of current clinical coagulation assays
Emicizumab interacts with all intrinsic pathway clotting-based labora-
tory assays, Table 3, which include all aPTT-based assays, such as the
OSA and activated clot time (ACT). The binding characteristics of emici-
zumab are such that even low plasma concentrations normalize the
aPTT, abolishing the typical relationship between FVIII activity and
aPTT used to monitor hemostatic effect. Under normal circumstances,
FIGURE 3 F8 pathogenic variants associated with alteration ofhydrophobic or charged amino acids decreasing the stability ofFVIII, resulting in discrepant FVIII activity reported by OSA andCSA. While the OSA is not sensitive to this reduced FVIII stabilitydue to the time course of the assay, the incubation time of theCSA permits more accurate assessment of FVIII activity anddetection of mild deficiency. FVIII molecule (with B domainremoved) with heavy chain (purple) and light chain (blue)illustrated; most mutations are clustered at A1-A2, A1-A3, and A2-A3 interfaces.12 The opposite relationship occurs for pathogenicvariants located in FIX- binding and thrombin cleavage sites (notshown in this figure); OSA is more sensitive to detecting resultantmild deficiencies compared to CSA. 3D render of FVIII protein cre-ated using Cn3D macromolecular structure viewer, NCBI StructureGroup59 [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 3 Coagulation assays to be avoided or used with caution inindividuals actively or recently receiving emicizumab66
Activated partial thromboplastin time (aPTT)One-stage clotting assay (aPTT-based)Bethesda or Nijmegen-Bethesda assay (aPTT-based)Activated clotting timeChromogenic substrate assay for FVIII(utilizing bovine protein reagents)Chromogenic substrate assay for FVIII(utilizing human protein reagents)Activated protein C resistance assay (aPTT-based)
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 6
6 | AJHAJH AL-SAMKARI AND CROTEAUAL-SAMKARI AND CROTEAU 1087
aPTT-based clotting assays measure the total clotting time of the
intrinsic pathway of coagulation, including activation of FVIII to FVIIIa
by thrombin; emicizumab does not require activation by thrombin and
will therefore result in a supra-physiologically short clotting time.16
Therefore, OSA performed in the presence of emicizumab would be
expected to grossly overestimate FVIII:C.7 While this is less clinically
relevant in the inhibitor population for whom laboratory monitoring of
FVIII:C is not generally clinically useful, standard inhibitor assays and
monitoring of use of FVIII in low-titer inhibitor patients with break-
through bleed events is not feasible with local laboratories.
5.3 | Quantifying hemostasis with emicizumab
The current FDA-approved CSA products commercially available in the
U.S. are unable to assay the hemostatic impact of emicizumab as they
use bovine coagulation proteins and thus are unable to detect emicizu-
mab, a humanized antibody.7 Therefore, while these assays cannot be
used to estimate emicizumab activity, they could be used to quantify
both exogenous and endogenous FVIII activity even in the presence of
emicizumab,16 for example in the noninhibitor patient population. Use
of emicizumab in this population will require thoughtful education of
not only hemophilia providers but also patients, families, and other
medical providers (particularly in operative and emergency settings)
and careful communication with the local coagulation laboratory about
available and appropriate assays.
In contrast, human CSA reagents are available in the research labo-
ratory setting and thus can provide a readout of FVIII activity that
includes the emicizumab contribution. Available data suggests the meas-
ured FVIII activity in patients on emicizumab overestimates the patient’s
actual hemostatic potential.16,67 With co-administration of emicizumab
and FVIII concentrates, hemostasis shortly after FVIII infusion are likely
to be driven by the exogenous FVIII because at high concentrations
FVIII will outcompete emicizumab for FIXa binding (FVIIIa binds FIXa
with KD of approximately 15 nM, in comparison with emicizumab, which
has a much lower affinity for FIXa, binding it with a KD of 1.5 mM).68 The
duration of this effect will depend on the individual’s FVIII clearance
which is highly variable among patients and often not known a priori.
6 | NOVEL APPROACHES TO ASSAYINGHEMOSTASIS
The global hemostasis assays including thrombin generation assays and
viscoelastic assays may ultimately emerge as an option for the monitor-
ing of hemophilia patients, particularly those treated with emicizumab
or other nonfactor hemostatic agents currently in early phase clinical
trials.69–71 The reported clinical utility of these assays varies in part due
to poor standardization of protocols and dependence on skill of individ-
uals performing the assays as well as the intended application of the
results. Novel strategies to capture hemostatic capacity in the broader
context of the vessel microenvironment incorporating the complexity
of interactions between endothelial cells, platelets, and coagulation fac-
tors are necessary.72 Microfluidic technologies may also provide an
opportunity for point of care testing and reduction of blood volume
required.73
7 | CONCLUSIONS
The aPTT-based OSA is the primary assay used for routine monitoring
and diagnosis of patients with hemophilia; however, use of CSA is gain-
ing in popularity given the expanding armamentarium of factor concen-
trates available and the numerous product-specific considerations to
keep in mind when using different analyzers and reagents for OSA.
First, some of the new bioengineered factor products and now the
novel hemostatic agents such as emicizumab cannot be accurately
measured by OSA and in some cases CSA. Knowing the specific situa-
tions in which OSA is unreliable and CSA should be used is critical for
the safe use of factor concentrates in the modern era. Dialogue to
share knowledge of aPTT reagents and patient factor products used
locally benefits both hemophilia providers and the coagulation labora-
tory team. Emicizumab is a breakthrough therapeutic in the manage-
ment of patients with hemophilia A, particularly those with inhibitors,
but it renders standard coagulation assays used in this population unre-
liable and therefore unusable for clinical care, particularly in emergent
clinical settings. As data emerge for potential use of emicizumab in the
noninhibitor hemophilia A population, recognition of its impact on rou-
tine clinical laboratories will become even more critical to ensure safe,
effective care. New assays or modifications of currently employed
assays are on the horizon, potentially including global hemostasis
assays and thrombin generation assays, to describe the hemostatic
potential of individual patients and optimize the management of hemo-
philia patients receiving novel therapeutics.
ACKNOWLEDGMENTS
The authors thank Dr. Maher Al-Samkari for graphic design assis-
tance with Figures 1 and 2.
AUTHOR CONTRIBUTIONS
H. Al-Samkari wrote the first draft and subsequent revisions of the
manuscript and contributed to concept and design, critical writing of
the intellectual content, and final approval; S.E. Croteau contributed
to concept and design, critical writing and revising the intellectual
content, and final approval.
ORCID
Hanny Al-Samkari http://orcid.org/0000-0001-6175-1383
Stacy E. Croteau http://orcid.org/0000-0001-6112-6166
REFERENCES
[1] Peters RT, Toby G, Lu Q, et al. Biochemical and functional charac-
terization of a recombinant monomeric factor VIII-Fc fusion protein.
J Thromb Haemost. 2013;11(1):132–141.
[2] Turecek PL, Bossard MJ, Graninger M, et al. BAX 855, a PEGylated
rFVIII product with prolonged half-life. Development, functional and
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 7
AL-SAMKARI AND CROTEAU AJHAJH | 71088 AL-SAMKARI AND CROTEAU
structural characterisation. Hamostaseologie. 2012;32(S 01):S29–S38.
[3] Morfini M. Pharmacokinetic drug evaluation of albutrepenonacog
alfa (CSL654) for the treatment of hemophilia. Expert Opin Drug
Metab Toxicol. 2016;12(11):1359–1365.
[4] Powell JS, Pasi KJ, Ragni MV, et al. Phase 3 study of recombinant
factor IX Fc fusion protein in hemophilia B. N Engl J Med. 2013;369
(24):2313–2323.
[5] Negrier C, Knobe K, Tiede A, et al. Enhanced pharmacokinetic proper-
ties of a glycoPEGylated recombinant factor IX: a first human dose
trial in patients with hemophilia B. Blood. 2011;118(10):2695–2701.
[6] Scott LJ, Kim ES. Emicizumab-kxwh: first global approval. Drugs.
2018;78(2):269–274.
[7] Oldenburg J, Mahlangu JN, Kim B, et al. Emicizumab prophylaxis in
hemophilia A with inhibitors. N Engl J Med. 2017;377(9):809–818.
[8] Chowdary P, Lethagen S, Friedrich U, et al. Safety and pharmacoki-
netics of anti-TFPI antibody (concizumab) in healthy volunteers and
patients with hemophilia: a randomized first human dose trial.
J Thromb Haemost. 2015;13(5):743–754.
[9] Pasi KJ, Rangarajan S, Georgiev P, et al. Targeting of antithrombin
in Hemophilia A or B with RNAi therapy. N Engl J Med. 2017;377
(9):819–828.
[10] Gale AJ, Bhat V, Pellequer JL, et al. Safety, stability and pharmaco-
kinetic properties of (super)Factor Va, a novel engineered coagula-
tion Factor V for treatment of severe bleeding. Pharm Res. 2016;33
(6):1517–1526.
[11] Barrowcliffe TW. Factor VIII and thrombin generation assays: rele-
vance to pharmacokinetic studies in haemophilia A. Haemophilia.
2006;12(s4):23–29.
[12] Oldenburg J, Pavlova A. Discrepancy between one-stage and chro-
mogenic factor VIII activity assay results can lead to misdiagnosis of
haemophilia A phenotype. Hamostaseologie. 2010;30(04):207–211.
[13] Hubbard AR, Sands D, Sandberg E, et al. A multi-centre collabora-
tive study on the potency estimation of ReFacto. Thromb Haemost.
2003;90(6):1088–1093.
[14] Peyvandi F, Oldenburg J, Friedman KD. A critical appraisal of one-
stage and chromogenic assays of factor VIII activity. J Thromb Hae-
most. 2016;14(2):248–261.
[15] Chitlur M, Young G. Global assays in hemophilia. Semin Hematol.
2016;53(1):40–45.
[16] Hemlibra (emicizumab-kxwh). Package insert. Genentech, Inc, South
San Francisco, CA; 2017.
[17] Srivastava A, Brewer AK, Mauser-Bunschoten EP, et al. Guidelines
for the management of hemophilia. Haemophilia. 2013;19(1):e1–47.
[18] Barrowcliffe TW. Monitoring haemophilia severity and treatment: new
or old laboratory tests?. Haemophilia. 2004;10(Suppl 4):109–114.
[19] Morfini M, Lee M, Messori A. The design and analysis of half-life
and recovery studies for factor VIII and factor IX. Factor VIII/Factor
IX Scientific and Standardization Committee of the International
Society for Thrombosis and Haemostasis. Thromb Haemost. 1991;66
(3):384–386.
[20] Iorio A, Blanchette V, Blatny J, et al. Estimating and interpreting the
pharmacokinetic profiles of individual patients with hemophilia A or
B using a population pharmacokinetic approach: communication
from the SSC of the ISTH. J Thromb Haemost. 2017;15(12):2461–2465.
[21] Croteau SE, Callaghan MU, Davis J, et al. Focusing in on use of
pharmacokinetic profiles in routine hemophilia care. Res Pract
Thromb Haemost. 2018;00:1–8.
[22] Potgieter JJ, Damgaard M, Hillarp A. One-stage vs. chromogenic
assays in haemophilia A. Eur J Haematol. 2015;94(Suppl 77):38–44.
[23] Dodt J, Hubbard AR, Wicks SJ, et al. Potency determination of fac-
tor VIII and factor IX for new product labelling and postinfusion
testing: challenges for caregivers and regulators. Haemophilia. 2015;
21(4):543–549.
[24] National Hemophilia Foundation Medical and Science Advisory Com-
mittee Statement Regarding Use of Various Clotting Factor Assays to
Monitor Factor Replacement Therapy. 2014;MASAC Document #228.
[25] Langdell RD, Wagner RH, Brinkhous KM. Effect of antihemophilic
factor on one-stage clotting tests; a presumptive test for hemophilia
and a simple one-stage antihemophilic factor assy procedure. J Lab
Clin Med. 1953;41:637–647.
[26] Over J. Methodology of the one-stage assay of Factor VIII (VIII:C).
Scand J Haematol Suppl. 1984;41:13–24.
[27] Verbruggen B, Meijer P, Novakova I, et al. Diagnosis of factor VIII
deficiency. Haemophilia. 2008;14(s3):76–82.
[28] Mackie I, Cooper P, Lawrie A, et al. Guidelines on the laboratory
aspects of assays used in haemostasis and thrombosis. Int Jnl Lab
Hem. 2013;35(1):1–13.
[29] Biggs R, Eveling J, Richards G. The assay of antihaemophilic-
globulin activity. Br J Haematol. 1955;1(1):20–34.
[30] Seghatchian MJ, Miller-Anderson M, Howarth DJ, et al. Proceed-
ings: Automated techniques for estimating factor VIII levels in con-
centrate preparations. Thromb Diath Haemorrh. 1975;34(2):611.
[31] Wagenvoord RJ, Hendrix HH, Hemker HC. Development of a sim-
ple chromogenic factor VIII assay for clinical use. Pathophysiol Hae-
mos Thromb. 1989;19(4):196–204.
[32] Rosen S. Assay of factor VIII:C with a chromogenic substrate. Scand
J Haematol Suppl. 1984;40:139–145.
[33] Barrowcliffe TW. Standardization of FVIII & FIX assays. Haemophilia.
2003;9(4):397–402.
[34] Mani H, Lindhoff-Last E. Main considerable factors for correct labo-
ratory test interpretation under DOA treatment. Thrombosis J.
2013;11(1):22.
[35] Rosen S, Andersson M, Blomback M, et al. Clinical application of a
chromogenic substrate method for determination of factor VIII
activity. Thromb Haemost. 1985;54(4):818–823.
[36] Barrowcliffe TW, Tubbs JE, Wong MY. Evaluation of factor VIII
deficient plasmas. Thromb Haemost. 1993;70(3):433–437.
[37] Rothschild C, Amiral J, Adam M, et al. Preparation of factor-VIII-
depleted plasma with antibodies and its use for the assay of factor
VIII. Haemostasis. 1990;20(6):321–328.
[38] St Ledger K, Feussner A, Kalina U, et al. International comparative
field study evaluating the assay performance of AFSTYLA in plasma
samples at clinical hemostasis laboratories. J Thromb Haemost.
2018;16(3):555–564.
[39] Sommer JM, Moore N, McGuffie-Valentine B, et al. Comparative
field study evaluating the activity of recombinant factor VIII Fc
fusion protein in plasma samples at clinical haemostasis laboratories.
Haemophilia. 2014;20(2):294–300.
[40] Kitchen S, Beckmann H, Katterle Y, et al. BAY 81–8973, a full-
length recombinant factor VIII: results from an International com-
parative laboratory field study. Haemophilia. 2016;22(3):e192–e199.
[41] Viuff D, Barrowcliffe T, Saugstrup T, et al. International comparative
field study of N8 evaluating factor VIII assay performance. Haemo-
philia. 2011;17(4):695–702.
[42] Hillarp A, Bowyer A, Ezban M, et al. Measuring FVIII activity of gly-
copegylated recombinant factor VIII, N8-GP, with commercially
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 8
8 | AJHAJH AL-SAMKARI AND CROTEAUAL-SAMKARI AND CROTEAU 1089
available one-stage clotting and chromogenic assay kits: a two-
centre study. Haemophilia. 2017;23(3):458–465.
[43] Pouplard C, Caron C, Aillaud MF, et al. The use of the new ReFacto
AF Laboratory Standard allows reliable measurement of FVIII:C lev-
els in ReFacto AF mock plasma samples by a one-stage clotting
assay. Haemophilia. 2011;17:e958–962.
[44] Sommer JM, Buyue Y, Bardan S, et al. Comparative field study:
impact of laboratory assay variability on the assessment of recombi-
nant factor IX Fc fusion protein (rFIXFc) activity. Thromb Haemost.
2014;112(5):932–940.
[45] St. Ledger K, Fuessner A, Kalina U, Jacobs I, Voigt C, Bensen-Ken-
nedy D. Performance of a recombinant fusion protein linking coagu-
lation factor IX with albumin (rFIX-FP) in the one–stage assay.
Haemophilia. 2016;22:60.
[46] Bowyer AE, Hillarp A, Ezban M, et al. Measuring factor IX activity
of nonacog beta pegol with commercially available one-stage clot-
ting and chromogenic assay kits: a two-center study. J Thromb Hae-
most. 2016;14(7):1428–1435.
[47] Sorensen MH, Andersen S, Ezban M. Factor IX-deficient plasma
spiked with N9-GP behaves similarly to N9-GP post-administration
clinical samples in N9-GP ELISA and FIX activity assays. Haemo-
philia. 2015;21(6):832–836.
[48] Rosen P, Rosen S, Ezban M, et al. Overestimation of N-
glycoPEGylated factor IX activity in a one-stage factor IX clotting
assay owing to silica-mediated premature conversion to activated
factor IX. J Thromb Haemost. 2016;14(7):1420–1427.
[49] Lee CA, Kessler CM, Varon D, et al. Discrepancies in potency
assessment of recombinant FVIII concentrates. Haemophilia. 1998;4
(4):634–640.
[50] Gu JM, Ramsey P, Evans V, et al. Evaluation of the activated partial
thromboplastin time assay for clinical monitoring of PEGylated
recombinant factor VIII (BAY 94–9027) for haemophilia A. Haemo-
philia. 2014;20(4):593–600.
[51] Barrowcliffe TW, Watton J, Harman A, et al. Potency of high purity
factor VIII concentrates. Lancet. 1990;336(8707):124.
[52] Hubbard AR, Weller LJ, Bevan SA. A survey of one-stage and chro-
mogenic potencies in therapeutic factor VIII concentrates. Br J Hae-
matol. 2002;117(1):247–248.
[53] Rodgers SE, Duncan EM, Barbulescu DM, et al. In vitro kinetics of
factor VIII activity in patients with mild haemophilia A and a dis-
crepancy between one-stage and two-stage factor VIII assay results.
Br J Haematol. 2007;136(1):138–145.
[54] Parquet-Gernez A, Mazurier C, Goudemand M. Functional and
immunological assays of FVIII in 133 haemophiliacs–characterizationof a subgroup of patients with mild haemophilia A and discrepancy
in 1- and 2-stage assays. Thromb Haemost. 1988;59(2):202–206.
[55] Lyall H, Hill M, Westby J, et al. Tyr346–>Cys mutation results
in factor VIII:C assay discrepancy and a normal bleeding phenotype - is
this mild haemophilia A?. Haemophilia. 2007;0(0):071124115610001–071124115610080.
[56] Trossaert M, Regnault V, Sigaud M, et al. Mild hemophilia A with
factor VIII assay discrepancy: using thrombin generation assay
to assess the bleeding phenotype. J Thromb Haemost. 2008;6(3):
486–493.
[57] Pavlova A, Delev D, Pezeshkpoor B, et al. Haemophilia A mutations
in patients with non-severe phenotype associated with a discrep-
ancy between one-stage and chromogenic factor VIII activity assays.
Thromb Haemost. 2014;111(05):851–861.
[58] Fay PJ. Factor VIII structure and function. Int J Hematol. 2006;83
(2):103–108.
[59] Wang Y, Geer LY, Chappey C, et al. Cn3D: sequence and structure
views for Entrez. Trends Biochem Sci. 2000;25(6):300–302.
[60] Hakeos WH, Miao H, Sirachainan N, et al. Hemophilia A mutationswithin the factor VIII A2-A3 subunit interface destabilize factor VIIIaand cause one-stage/two-stage activity discrepancy. Thromb Hae-most. 2002;88(5):781–787.
[61] Bowyer A, Key N, Dalton D, et al. The coagulation laboratory moni-
toring of Afstyla single-chain FVIII concentrate. Haemophilia. 2017;23(5):e469–e470.
[62] Friedman KD, Powell JS, Bensen-Kennedy D. Response: The coagu-lation laboratory monitoring of AFSTYLA single-chain FVIII concen-trate. Haemophilia. 2018;24(3):e129–131.
[63] Hartmann J, Croteau SE. 2017 Clinical trials update: Innovations in
hemophilia therapy. Am J Hematol. 2016;91(12):1252–1260.
[64] Jimenez-Yuste V, Shima M, Fukutake K, et al. Emicizumab subcuta-neous dosing every 4 weeks for the management of Hemophilia A:preliminary data from the pharmacokinetic run-in cohort of a multi-center, open-label, phase 3 Study (HAVEN. Blood. 2017;130:486
[65] Muto A, Yoshihashi K, Takeda M, et al. Anti-factor IXa/X bispecificantibody (ACE910): hemostatic potency against ongoing bleeds in ahemophilia A model and the possibility of routine supplementation.J Thromb Haemost. 2014;12(2):206–213.
[66] Shima M, Hanabusa H, Taki M, et al. Factor VIII-mimetic function of
humanized bispecific antibody in Hemophilia A. N Engl J Med. 2016;374(21):2044–2053.
[67] Shima M, Hanabusa H, Taki M, et al. Long-term safety and effi-cacy of emicizumab in a phase 1/2 study in patients with hemo-philia A with or without inhibitors. Blood Adv. 2017;1(22):1891–1899.
[68] Lenting PJ, Denis CV, Christophe OD. Emicizumab, a bispecificantibody recognizing coagulation factors IX and X: how does itactually compare to factor VIII?. Blood. 2017;130(23):2463–2468.
[69] Young G, Sorensen B, Dargaud Y, et al. Thrombin generation and
whole blood viscoelastic assays in the management of hemophilia:current state of art and future perspectives. Blood. 2013;121(11):1944–1950.
[70] Salvagno GL, Berntorp E. Thrombin Generation Assays (TGAs).Methods Mol Biol. 2017;1646:515–522.
[71] Panteleev MA, Hemker HC. Global/integral assays in hemostasis
diagnostics: promises, successes, problems and prospects. Thrombo-sis J. 2015;13(1):5.
[72] Sakurai Y, Hardy ET, Ahn B, et al. A microengineered vascularizedbleeding model that integrates the principal components of hemo-stasis. Nat Commun. 2018;9(1):509.
[73] Padovani F, Duffy J, Hegner M. Nanomechanical clinical coagulationdiagnostics and monitoring of therapies. Nanoscale. 2017;9(45):17939–17947.
How to cite this article: Al-Samkari H, Croteau SE. Shifting
landscape of hemophilia therapy: Implications for current clinical
laboratory coagulation assays. Am J Hematol. 2018;00:1–9.
https://doi.org/10.1002/ajh.25153
J_ID: Customer A_ID: AJH25153 Cadmus Art: AJH25153 Ed. Ref. No.: AJH-18-0433.R1 Date: 3-July-18 Stage: Page: 9
AL-SAMKARI AND CROTEAU AJHAJH | 9
landscape of hemophilia therapy: Implications for current clinical
laboratory coagulation assays. Am J Hematol. 2018;93:1082–1090.
https://doi.org/10.1002/ajh.25153
1090 AL-SAMKARI AND CROTEAU